基于新型Top-hat和小波变换的红外弱小目标检测OA

According to the characteristics of infrared dim small target image,such as low signal strength,small size,susceptible to noise interference and complex background clutter,this paper proposes a new infrared dim and small target detection method based on the combination of a new morphological Top-hat operator and wavelet transform.Firstly,the improved morphological filtering was used to suppress the background of the infrared image,so as to reduce the interference of clouds,ground cover fluctuation and local strong clutter on target detection.Secondly,the preprocessed image was decomposed by wavelet,and the low frequency component was used for background suppression and target enhancement by using the Local Gradient and Local Intensity algorithm.At the same time,the three high frequency components were accumulated to further enhance the feature information of the target point.Finally,the infrared dim and small target was extracted by adaptive threshold segmentation method.The experimental results show that the proposed method achieves the optimal Signal-to-Clutter Ratio Gain(SCRG),second only to the PTCTV algorithm in the Background Suppression Factor(BSF),and the overall performance is better than other comparison methods.Among them,SCRG and BSF on Seq6 are 14.16 and 82.49,respectively.In the ROC curve,the detection rate of the proposed method in Seq3,Seq5 and Seq6 is more than 95%,which fully verifies that the algorithm can effectively suppress background and noise in complex scenes,and has strong stability and robustness.